Re: Performance improvement with Akira Tsukamoto's Athlon copy_user patch

From: Akira Tsukamoto (akira-t@suna-asobi.com)
Date: Wed Nov 20 2002 - 01:03:11 EST


Not deeply but I looked your exception safe kfpu.

'kernel_fpu_begin/end' was like delayed FPU sate switching,
on the other hand, kfpu is more abstracted FPU semaphore
which works per cpu.

I have not compiled it but it should work fine on 2.4, but
have some questions using kfpu for 2.5.

 1)Is 'per CPU' safe for 2.5?
   I think the new O(1) scheduler has process migration between cpus
   and I am a bit afraid.
   If not safe, any good idea for it?
   May be per task could be safer even not optimal for smp?
 2)Need to add preempt_disable/enable().
 3)struct kfpuacquire {} kfpuacquire; is in the kfpu.h header,
   might cause compile problem.

> Have you tried SSE based copy_to_user? With SSE, you can just save the affected registers, without unexpected sideeffects.

No, my CPU don't run SSE......:)

By the way,
do you have separeted patch only for kfpu?

Thank you,

Akira Tsukamoto

On Tue, 19 Nov 2002 20:18:25 +0100
Manfred Spraul <manfred@colorfullife.com> mentioned:

> >I read the code of laze FPU state saving and confirmed that
> >if the function does not generate exception than
> >'kernel_fpu_begin/end()' should assure fpu safe inside kernel.
> >
> >However, it is not enough where exception could rise, as Takahashi
> >mentioned.
>
>
> I had prototyped an exception safe kfpu framework, but then I didn't have the time to submit/cleanup it.
>
> http://www.colorfullife.com/~manfred/linux-2.5/sse/patch-kfpu
>
> Have you tried SSE based copy_to_user? With SSE, you can just save the affected registers, without unexpected sideeffects.
>
> --
> Manfred
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/

-- 
Akira Tsukamoto <akira-t@suna-asobi.com, at541@columbia.edu>

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sat Nov 23 2002 - 22:00:30 EST